INDEX

Explanations

quotations and dialogue

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.78

</h3>

-0.77

her

-0.68

da

-0.68

van

-0.67

</h6>

-0.67

«

-0.67

et

-0.66

im

-0.64

POSITIVE LOGITS

 quæ

1.17

 Reſ

1.13

 ſever

1.13

 Perſ

1.12

 ſtate

1.12

 Anſ

1.10

 reaſon

1.10

 itſelf

1.10

 myſelf

1.09

 Diſ

1.09

Activations Density 0.137%

quotations and dialogue

No Comments

No Known Activations